WO2022227202A1 - 一种运动状态识别方法、系统、可穿戴设备和存储介质 - Google Patents
一种运动状态识别方法、系统、可穿戴设备和存储介质 Download PDFInfo
- Publication number
- WO2022227202A1 WO2022227202A1 PCT/CN2021/097181 CN2021097181W WO2022227202A1 WO 2022227202 A1 WO2022227202 A1 WO 2022227202A1 CN 2021097181 W CN2021097181 W CN 2021097181W WO 2022227202 A1 WO2022227202 A1 WO 2022227202A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- target
- motion state
- decision
- resultant acceleration
- acceleration
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 230000001133 acceleration Effects 0.000 claims abstract description 313
- 238000003066 decision tree Methods 0.000 claims abstract description 73
- 238000004891 communication Methods 0.000 claims description 28
- 238000012544 monitoring process Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 7
- 230000009471 action Effects 0.000 claims description 5
- 230000008569 process Effects 0.000 abstract description 11
- 230000006870 function Effects 0.000 description 11
- 239000006185 dispersion Substances 0.000 description 9
- 238000012545 processing Methods 0.000 description 7
- 238000012549 training Methods 0.000 description 6
- 239000000284 extract Substances 0.000 description 5
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 1
- 230000010267 cellular communication Effects 0.000 description 1
- 230000001351 cycling effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/10—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
- G01C21/12—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
- G01C21/16—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01P—MEASURING LINEAR OR ANGULAR SPEED, ACCELERATION, DECELERATION, OR SHOCK; INDICATING PRESENCE, ABSENCE, OR DIRECTION, OF MOVEMENT
- G01P15/00—Measuring acceleration; Measuring deceleration; Measuring shock, i.e. sudden change of acceleration
- G01P15/18—Measuring acceleration; Measuring deceleration; Measuring shock, i.e. sudden change of acceleration in two or more dimensions
Definitions
- the patent of the present invention relates to the field of identification technology, in particular to a motion state identification method, system, wearable device and storage medium.
- the monitoring terminal obtains the positioning position, it also pays more attention to the current motion state of the target to judge its safety situation.
- the motion states that we can identify are only walking, running, and cycling.
- the motion states of the target are increasingly diverse, so there is a strong demand for stable identification of different motion states.
- the recognition of different motion states relies on the speed information provided by GPS, which consumes a lot of power. In many scenarios, such as tunnels or subways, there will be no GPS signal, and there will be no recognition.
- the wearable device compares the motion parameters and state thresholds at a time, and judges that the motion state is unstable. When reporting the motion state to the monitoring terminal, it is prone to misjudgment of the motion state.
- the present invention provides a motion state identification method, system, wearable device and storage medium.
- the characteristic values of the resultant acceleration including: mean, variance, and kurtosis
- they are input into a decision tree model, and according to the decision
- the motion state of the target is judged, which solves the problems in the prior art that the motion state of the target cannot be accurately recognized when the GPS signal is too weak, and the problem of misjudgment in the process of recognizing the motion state is solved.
- the accuracy of identifying the motion state of the target is improved, and the influence of interference information on the motion state of the target received by the monitoring terminal is reduced.
- the present invention provides a motion state identification method, comprising:
- Extracting the characteristic value of the resultant acceleration of the target includes: mean, variance, and kurtosis;
- the current motion state of the target is judged according to the decision result.
- the technical solution can solve the problem of how to identify the motion state of the target by measuring the acceleration of the target when the GPS positioning is inaccurate or the GPS signal is not good.
- the eigenvalues of the target resultant acceleration including: mean, variance and kurtosis
- the current motion state of the target can be accurately identified, using the mean, variance, and kurtosis of the target resultant acceleration.
- the unique advantages of kurtosis and other eigenvalues to distinguish different motion states can effectively reduce the error in the identification process.
- a kind of motion state identification method also comprises:
- the statistics of the several decision-making results in a unit time are obtained to obtain the current motion state of the target in a unit time, which specifically includes:
- a kind of motion state identification method also comprises:
- the information that the target is currently in a certain motion state is not output.
- a kind of motion state identification method also comprises:
- the present invention also provides a motion state identification system, comprising:
- the acquisition module is used to collect the three-dimensional acceleration data of the target
- an arithmetic module for calculating the resultant acceleration of the target according to the three-dimensional acceleration data of the target
- an extraction module configured to extract the characteristic value of the resultant acceleration of the target, where the characteristic value of the resultant acceleration includes: mean, variance, and kurtosis;
- a decision tree module for inputting the characteristic value of the resultant acceleration into a preset decision tree model to obtain a decision result
- a statistical calculation module for judging the current motion state of the target according to the decision result
- a communication module used for outputting the information of the current motion state of the target within the unit time
- the monitoring terminal is configured to receive the information of the current motion state of the target within the unit time.
- the statistical calculation module includes:
- the communication module includes:
- the present invention also provides a wearable device, comprising an acceleration sensor, a processor, a storage, and a computer program stored in the storage and executable on the processor, the processor for executing the storage
- the computer program stored on the computer program implements the operations performed by the motion state identification method according to any one of claims 1 to 5.
- the present invention further provides a storage medium, characterized in that, the storage medium stores at least one instruction, and the instruction is loaded and executed by a processor to implement the motion according to any one of claims 1 to 5 The action performed by the state identification method.
- the present invention provides a motion state identification method, system, wearable device and storage medium, which at least have the following gain effects:
- Fig. 1 is a flow chart of an embodiment of a motion state identification method of the present invention
- FIG. 2 is a flowchart of an embodiment of a decision tree in a motion state identification method of the present invention
- FIG. 3 is a flowchart of another embodiment of a motion state identification method of the present invention.
- FIG. 4 is a flowchart of another embodiment of a motion state identification method of the present invention.
- Fig. 5 is a flow chart of an embodiment of a motion state identification system of the present invention.
- FIG. 6 is a schematic structural diagram of an embodiment of a wearable device of the present invention.
- Labels in the figure 10-collection module, 20-operation module, 30-extraction module, 40-decision tree module, 50-statistics module, 60-communication module, 70-monitoring terminal, 100-wearable device, 110-acceleration Sensor, 120-memory, 121-computer program, 130-processor.
- An embodiment of the present invention is a method for identifying a motion state, including:
- S100 collects the three-dimensional acceleration data of the target
- collecting the three-dimensional acceleration data of the target refers to collecting the accelerations a x , a y , and az on the three direction axes of the acceleration sensor x, y, and z of the target.
- S200 calculates the resultant acceleration of the target according to the three-dimensional acceleration data of the target
- the three-dimensional acceleration data a x , a y , and az of the target are collected. Considering the variety of wearable devices and actions, it is not meaningful to extract the acceleration features separately for the three direction axes. Merge, extract the vector sum of the accelerations a x , a y , and a z on the three direction axes as the resultant acceleration a, and then perform feature extraction on the resultant acceleration a.
- the resultant acceleration a is calculated as:
- S300 extracts the characteristic value of the resultant acceleration of the target, and the characteristic value of the resultant acceleration includes: mean, variance, and kurtosis;
- the mean value in the extracted feature values is the average amount of acceleration representing the target in a unit time.
- the mean value can be used to distinguish different motion states of the target, such as walking state, running state,
- the mean value of acceleration is different in the state of taking the car, the state of taking the subway, and the state of taking the bus, and the target is in different motion states;
- the variance in the extracted eigenvalues is to characterize the acceleration dispersion degree of the target in unit time.
- the target When the target is in motion states with different acceleration dispersion degrees, it can also be used to distinguish different motion states of the target, such as the unit time in the uniform walking motion state.
- the discrete degree of the internal acceleration is low.
- the variance can also be used to reduce the error of identifying the target motion state. The calculation of the variance is:
- the kurtosis in the extracted eigenvalues is to characterize the acceleration of the target in unit time and the state of the probability distribution density function curve at the average value. By judging the kurtosis, the error of identifying the motion state of the target can be reduced, and the accuracy of the motion state can be improved. Accuracy of identification, where kurtosis is calculated as:
- the skewness of the resultant acceleration may also be extracted.
- the skewness in the extracted eigenvalues is the asymmetry of the probability distribution density function curve of the target in unit time relative to the average value.
- S400 inputs the characteristic value of the resultant acceleration into the preset decision tree model to obtain the decision result
- the steps for establishing the preset decision tree model are:
- the extracted eigenvalues of the acceleration are input into the decision tree model for discrimination, and the parameters in the decision tree model are trained from more than 12 hours of test data.
- the detailed judgment process for obtaining the decision result includes:
- Input the eigenvalue of the resultant acceleration into the decision tree model when the variance value of the resultant acceleration in the first decision is greater than the first variance value, enter the second decision, when the variance value of the resultant acceleration is less than the second variance value , enter the third decision, when the kurtosis of the resultant acceleration is greater than the first kurtosis, enter the fourth decision, when the mean of the resultant acceleration is greater than the second mean, the output decision result is A;
- Input the eigenvalue of the resultant acceleration into the decision tree model when the variance value of the resultant acceleration in the first decision is greater than the first variance value, enter the second decision, when the variance value of the resultant acceleration is less than the second variance value , enter the third decision, when the kurtosis of the resultant acceleration is greater than the first kurtosis, enter the fourth decision, when the mean of the resultant acceleration is less than the second mean, enter the fifth decision, when the variance is less than the third variance value, the output decision result is B;
- Input the eigenvalue of the resultant acceleration into the decision tree model when the variance value of the resultant acceleration in the first decision is greater than the first variance value, enter the second decision, when the variance value of the resultant acceleration is less than the second variance value , enter the third decision, when the kurtosis of the resultant acceleration is greater than the first kurtosis, enter the fourth decision, when the mean of the resultant acceleration is less than the second mean, enter the fifth decision, when the variance of the resultant acceleration is greater than When the third difference is reached, the sixth decision is entered.
- the output decision result is A;
- Input the eigenvalue of the resultant acceleration into the decision tree model when the variance value of the resultant acceleration in the first decision is greater than the first variance value, enter the second decision, when the variance value of the resultant acceleration is greater than the second variance value , enter the third decision, when the kurtosis of the resultant acceleration is less than the second kurtosis, enter the fourth decision, when the mean of the resultant acceleration is less than the third mean, enter the fifth decision, when the variance of the resultant acceleration is less than the third decision
- the output decision result is C;
- Input the eigenvalue of the resultant acceleration into the decision tree model when the variance value of the resultant acceleration in the first decision is greater than the first variance value, enter the second decision, when the variance value of the resultant acceleration is greater than the second variance value , enter the third decision, when the kurtosis of the resultant acceleration is less than the second kurtosis, enter the fourth decision, when the mean of the resultant acceleration is less than the third mean, enter the fifth decision, when the variance of the resultant acceleration is greater than the third decision
- the output decision result is B.
- S500 judges the current motion state of the target according to the decision result.
- This embodiment can solve the problem of how to identify the motion state of the target by measuring the acceleration of the target when the GPS positioning is inaccurate or the GPS signal is not good.
- the eigenvalues of the target resultant acceleration including: mean, variance and kurtosis
- the current motion state of the target can be accurately identified, using the mean, variance, and kurtosis of the target resultant acceleration.
- the unique advantages of kurtosis and other eigenvalues to distinguish different motion states can effectively reduce the error in the identification process.
- FIG. 3 Another embodiment of the present invention, as shown in FIG. 3 , is a method for identifying a motion state, including:
- S100 collects the three-dimensional acceleration data of the target
- collecting the three-dimensional acceleration data of the target refers to collecting the accelerations a x , a y , and az on the three direction axes of the acceleration sensor x, y, and z of the target.
- S200 calculates the resultant acceleration of the target according to the three-dimensional acceleration data of the target
- the three-dimensional acceleration data a x , a y , and az of the target are collected. Considering the variety of wearable devices and actions, it is not meaningful to extract the acceleration features separately for the three direction axes. Merge, extract the vector sum of the accelerations a x , a y , and a z on the three direction axes as the resultant acceleration a, and then perform feature extraction on the resultant acceleration a.
- the resultant acceleration a is calculated as:
- S300 extracts the characteristic value of the resultant acceleration of the target, and the characteristic value of the resultant acceleration includes: mean, variance, and kurtosis;
- the mean value in the extracted feature values is the average amount of acceleration representing the target in a unit time.
- the mean value can be used to distinguish different motion states of the target, such as walking state, running state,
- the mean value of acceleration is different in the state of taking the car, the state of taking the subway, and the state of taking the bus, and the target is in different motion states;
- the variance in the extracted eigenvalues is to characterize the acceleration dispersion degree of the target in unit time.
- the target When the target is in motion states with different acceleration dispersion degrees, it can also be used to distinguish different motion states of the target, such as the unit time in the uniform walking motion state.
- the discrete degree of the internal acceleration is low.
- the variance can also be used to reduce the error of identifying the target motion state. The calculation of the variance is:
- the kurtosis in the extracted eigenvalues is to characterize the acceleration of the target in unit time and the state of the probability distribution density function curve at the average value. By judging the kurtosis, the error of identifying the motion state of the target can be reduced, and the accuracy of the motion state can be improved. Accuracy of identification, where kurtosis is calculated as:
- the skewness of the resultant acceleration may also be extracted.
- the skewness in the extracted eigenvalues is the asymmetry of the probability distribution density function curve of the target in unit time relative to the average value.
- S400 inputs the characteristic value of the resultant acceleration into the preset decision tree model to obtain the decision result
- the steps for establishing the preset decision tree model are:
- the extracted eigenvalues of the acceleration are input into the decision tree model for discrimination, and the parameters in the decision tree model are trained from more than 12 hours of test data.
- the detailed judgment process for obtaining the decision result includes:
- Input the eigenvalue of the resultant acceleration into the decision tree model when the variance value of the resultant acceleration in the first decision is greater than the first variance value, enter the second decision, when the variance value of the resultant acceleration is less than the second variance value , enter the third decision, when the kurtosis of the resultant acceleration is greater than the first kurtosis, enter the fourth decision, when the mean value of the resultant acceleration is greater than the second mean, the output decision result is A;
- Input the eigenvalue of the resultant acceleration into the decision tree model when the variance value of the resultant acceleration in the first decision is greater than the first variance value, enter the second decision, when the variance value of the resultant acceleration is less than the second variance value , enter the third decision, when the kurtosis of the resultant acceleration is greater than the first kurtosis, enter the fourth decision, when the mean of the resultant acceleration is less than the second mean, enter the fifth decision, when the variance is less than the third variance value, the output decision result is B;
- Input the eigenvalue of the resultant acceleration into the decision tree model when the variance value of the resultant acceleration in the first decision is greater than the first variance value, enter the second decision, when the variance value of the resultant acceleration is less than the second variance value , enter the third decision, when the kurtosis of the resultant acceleration is greater than the first kurtosis, enter the fourth decision, when the mean of the resultant acceleration is less than the second mean, enter the fifth decision, when the variance of the resultant acceleration is greater than When the third difference is reached, the sixth decision is entered.
- the output decision result is A;
- Input the eigenvalue of the resultant acceleration into the decision tree model when the variance value of the resultant acceleration in the first decision is greater than the first variance value, enter the second decision, when the variance value of the resultant acceleration is greater than the second variance value , enter the third decision, when the kurtosis of the resultant acceleration is less than the second kurtosis, enter the fourth decision, when the mean of the resultant acceleration is less than the third mean, enter the fifth decision, when the variance of the resultant acceleration is less than the third decision
- the output decision result is C;
- Input the eigenvalue of the resultant acceleration into the decision tree model when the variance value of the resultant acceleration in the first decision is greater than the first variance value, enter the second decision, when the variance value of the resultant acceleration is greater than the second variance value , enter the third decision, when the kurtosis of the resultant acceleration is less than the second kurtosis, enter the fourth decision, when the mean of the resultant acceleration is less than the third mean, enter the fifth decision, when the variance of the resultant acceleration is greater than the third decision
- the output decision result is B.
- S510 Counting several decision results in a unit time, when the proportion of the number of decision results corresponding to a certain motion state of the target reaches a preset threshold of the total number of several decision results of the target, then determine the motion The state is the current motion state of the target.
- the preset threshold can be set between 50% and 90%.
- the preset threshold is set at 80%, that is, when the number of decision results corresponding to a certain motion state of the target accounts for several decision results of the target
- the motion state is determined to be the current motion state of the target.
- FIG. 4 Another embodiment of the present invention, as shown in FIG. 4 , is based on the previous embodiment, a motion state identification method, further comprising:
- the communication module when it is judged that the accuracy of the current motion state of the target is not high, the communication module does not output the information of the current motion state of the target to the monitoring terminal, which improves the accuracy of the monitoring terminal acquiring the state information.
- the present invention also provides a motion state recognition system, including:
- the acquisition module 10 is used for acquiring the three-dimensional acceleration data of the target
- collecting the three-dimensional acceleration data of the target refers to collecting the accelerations a x , a y , and az on the three direction axes of the acceleration sensor x, y, and z of the target.
- an arithmetic module 20 configured to calculate the resultant acceleration of the target according to the three-dimensional acceleration data of the target;
- the three-dimensional acceleration data a x , a y , and az of the target are collected. Considering the variety of wearable devices and actions, it is not meaningful to extract the acceleration features separately for the three direction axes. Merge, extract the vector sum of the accelerations a x , a y , and a z on the three direction axes as the resultant acceleration a, and then perform feature extraction on the resultant acceleration a.
- the resultant acceleration a is calculated as:
- the extraction module 30 is configured to extract the characteristic value of the resultant acceleration of the target, where the characteristic value of the resultant acceleration includes: mean, variance, and kurtosis;
- the mean value in the extracted feature values is the average amount of acceleration representing the target in a unit time.
- the mean value can be used to distinguish different motion states of the target, such as walking state, running state,
- the mean value of acceleration is different in the state of taking the car, the state of taking the subway, and the state of taking the bus, and the target is in different motion states;
- the variance in the extracted eigenvalues is to characterize the acceleration dispersion degree of the target in unit time.
- the target When the target is in motion states with different acceleration dispersion degrees, it can also be used to distinguish different motion states of the target, such as the unit time in the uniform walking motion state.
- the discrete degree of the internal acceleration is low.
- the variance can also be used to reduce the error of identifying the target motion state. The calculation of the variance is:
- the kurtosis in the extracted eigenvalues is to characterize the acceleration of the target in unit time and the state of the probability distribution density function curve at the average value. By judging the kurtosis, the error of identifying the motion state of the target can be reduced, and the accuracy of the motion state can be improved. Accuracy of identification, where kurtosis is calculated as:
- the skewness of the resultant acceleration may also be extracted.
- the skewness in the extracted eigenvalues is the asymmetry of the probability distribution density function curve of the target in unit time relative to the average value.
- a decision tree module 40 configured to input the characteristic value of the resultant acceleration into a preset decision tree model to obtain a decision result
- the steps for establishing the preset decision tree model are:
- the extracted eigenvalues of the acceleration are input into the decision tree model for discrimination, and the parameters in the decision tree model are trained from more than 12 hours of test data.
- the detailed judgment process for obtaining the decision result includes:
- Input the eigenvalue of the resultant acceleration into the decision tree model when the variance value of the resultant acceleration in the first decision is greater than the first variance value, enter the second decision, when the variance value of the resultant acceleration is less than the second variance value , enter the third decision, when the kurtosis of the resultant acceleration is greater than the first kurtosis, enter the fourth decision, when the mean of the resultant acceleration is greater than the second mean, the output decision result is A;
- Input the eigenvalue of the resultant acceleration into the decision tree model when the variance value of the resultant acceleration in the first decision is greater than the first variance value, enter the second decision, when the variance value of the resultant acceleration is less than the second variance value , enter the third decision, when the kurtosis of the resultant acceleration is greater than the first kurtosis, enter the fourth decision, when the mean of the resultant acceleration is less than the second mean, enter the fifth decision, when the variance is less than the third variance value, the output decision result is B;
- Input the eigenvalue of the resultant acceleration into the decision tree model when the variance value of the resultant acceleration in the first decision is greater than the first variance value, enter the second decision, when the variance value of the resultant acceleration is less than the second variance value , enter the third decision, when the kurtosis of the resultant acceleration is greater than the first kurtosis, enter the fourth decision, when the mean of the resultant acceleration is less than the second mean, enter the fifth decision, when the variance of the resultant acceleration is greater than When the third difference is reached, the sixth decision is entered.
- the output decision result is A;
- Input the eigenvalue of the resultant acceleration into the decision tree model when the variance value of the resultant acceleration in the first decision is greater than the first variance value, enter the second decision, when the variance value of the resultant acceleration is greater than the second variance value , enter the third decision, when the kurtosis of the resultant acceleration is less than the second kurtosis, enter the fourth decision, when the mean of the resultant acceleration is less than the third mean, enter the fifth decision, when the variance of the resultant acceleration is less than the third decision
- the output decision result is C;
- Input the eigenvalue of the resultant acceleration into the decision tree model when the variance value of the resultant acceleration in the first decision is greater than the first variance value, enter the second decision, when the variance value of the resultant acceleration is greater than the second variance value , enter the third decision, when the kurtosis of the resultant acceleration is less than the second kurtosis, enter the fourth decision, when the mean of the resultant acceleration is less than the third mean, enter the fifth decision, when the variance of the resultant acceleration is greater than the third decision
- the output decision result is B.
- the statistical calculation module 50 is used for judging the current motion state of the target according to the decision result
- the communication module 60 is used for outputting the information of the current motion state of the target in unit time
- the monitoring terminal 70 is used for receiving the information of the current motion state of the target within a unit time.
- This embodiment can solve the problem of how to identify the motion state of the target by measuring the acceleration of the target when the GPS positioning is inaccurate or the GPS signal is not good.
- the eigenvalues of the target resultant acceleration including: mean, variance and kurtosis
- the current motion state of the target can be accurately identified, using the mean, variance, and kurtosis of the target resultant acceleration.
- the unique advantages of kurtosis and other eigenvalues to distinguish different motion states can effectively reduce the error in the identification process.
- the statistical calculation module further includes the following functions:
- the preset threshold can be set between 50% and 90%.
- the preset threshold is set at 80%, that is, when the number of several decision results corresponding to a certain motion state of the target accounts for several of the target
- the total number of decision-making results is more than 80%, it is judged that the target is currently in the motion state; when the number of decision-making results corresponding to a certain motion state of the target accounts for 80% of the total number of decision-making results of the target In the following cases, it is determined that the target is not currently in the motion state.
- this module it is specifically explained how to judge the current motion state of the target according to several decision results in a unit time. Through the design of preset thresholds, the error generated in the process of identifying the motion state of the target is reduced, and the motion state recognition is improved. accuracy.
- the communication module further includes the following functions:
- the communication module when it is judged that the accuracy of the current motion state of the target is not high, the communication module does not output the information of the current motion state of the target to the monitoring terminal, which improves the accuracy of the monitoring terminal acquiring the status information.
- the present invention further provides a wearable device 100 , which includes an acceleration sensor 110 and a memory 120 for storing a computer program 121 and a processor 130 for executing all data stored in the memory 120 .
- the stored computer program 121 implements the method for recognizing the motion state in the corresponding method embodiment above.
- the wearable device 100 may be a desktop computer, a notebook, a palmtop computer, a tablet computer, a mobile phone, a human-computer interaction screen, and other devices.
- the wearable device 100 may include, but is not limited to, the processor 130 and the memory 120 .
- FIG. 6 is only an example of the wearable device 100, and does not constitute a limitation on the wearable device 100, and may include more or less components than the one shown, or combine some components, or different Components, for example, the wearable device 100 may also include input/output interfaces, display devices, network access devices, communication buses, communication interfaces, and the like.
- the communication interface and the communication bus may also include an input/output interface, wherein the processor 130, the memory 120, the input/output interface and the communication interface communicate with each other through the communication bus.
- the memory 120 stores a computer program 121
- the processor 130 is configured to execute the computer program 121 stored in the memory 120 to implement the photographing control method in the corresponding method embodiment above.
- the processor 130 may be a central processing unit (Central Processing Unit, CPU), or other general-purpose processors, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field-available processor. Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
- a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
- the memory 120 may be an internal storage unit of the wearable device 100, such as a hard disk or a memory of the wearable device.
- the memory can also be an external storage device of the wearable device, for example: a pluggable hard disk equipped on the wearable device, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash memory card ( Flash Card), etc.
- the memory 120 may also include both an internal storage unit of the wearable device 100 and an external storage device.
- the memory 120 is used to store the computer program 121 and other programs and data required by the wearable device 100 . Wearable devices can also be used to temporarily store data that has been or will be output.
- a communication bus is a circuit that connects the described elements and enables transmission between these elements.
- the processor 130 receives commands from other elements through a communication bus, decrypts the received commands, and performs computation or data processing according to the decrypted commands.
- the memory 120 may include program modules such as a kernel (kernel), middleware (middleware), application programming interface (Application Programming Interface, API) and applications.
- the program module may be composed of software, firmware or hardware, or at least two of them.
- the input/output interface forwards commands or data entered by the user through the input/output interface (eg, sensor, keyboard, touch screen).
- the communication interface connects the wearable device 100 with other network devices, user equipment, and networks.
- the communication interface may be wired or wirelessly connected to a network to connect to external other network devices or user equipment.
- Wireless communication may include at least one of: Wireless Fidelity (WiFi), Bluetooth (BT), Near Field Communication Technology (NFC), Global Positioning System (GPS), cellular communication, and the like.
- the wired communication may include at least one of: Universal Serial Bus (USB), High Definition Multimedia Interface (HDMI), Asynchronous Transfer Standard Interface (RS-232), and the like.
- the network may be a telecommunication network and a communication network.
- the communication network can be a computer network, the Internet, the Internet of Things, a telephone network.
- the wearable device 100 may be connected to a network through a communication interface, and the protocol used by the wearable device 100 to communicate with other network devices may be supported by at least one of an application, an application programming interface (API), a middleware, a kernel, and a communication interface.
- API application programming interface
- the present invention also provides a storage medium, where at least one instruction is stored, and the instruction is loaded and executed by a processor to implement the operations performed by the corresponding embodiments of the above-mentioned photographing control method.
- the storage medium may be read only memory (ROM), random access memory (RAM), compact disk read only (CD-ROM), magnetic tapes, floppy disks, optical data storage devices, and the like.
- the disclosed device/smart watch and method may be implemented in other ways.
- the device/smart watch embodiments described above are only illustrative.
- the division of the modules or units is only a logical function division.
- there may be other division methods for example, a plurality of Units or components may be combined or may be integrated into another system, or some features may be omitted, or not implemented.
- the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, which may be in electrical, mechanical or other forms.
- the units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.
- each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.
- the above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
一种运动状态识别方法、系统、可穿戴设备和存储介质，其方法包括：采集目标的三维加速度数据（S100）；根据该目标的三维加速度数据，计算该目标的合加速度（S200）；提取该目标的合加速度的特征值，该合加速度的特征值包括：均值、方差、峰度（S300）；将该合加速度的特征值输入预设决策树模型，获得决策结果（S400）；根据该决策结果判断该目标的当前运动状态（S500）。该方法针对当GPS信号过弱时无法准确识别目标运动状态、以及在识别运动状态过程中误判的问题，有效提高了识别目标运动状态的准确性。
Description
本发明专利涉及识别技术领域，尤其指一种运动状态识别方法、系统、可穿戴设备和存储介质。
对于需要识别运动状态的目标而言，监控终端在获取定位位置的时候，同样比较关心目标当前的运动状态，用来判断其安全情况，通常我们能识别的运动状态只有，走路，跑步，骑车等特征比较明显的状态，然而现在目标的运动状态日益多样性，因此也出现了对不同运动状态的稳定识别的强烈需求。
目前对于不同运动状态的识别，依赖于GPS提供的速度信息，功耗很大，而且在很多场景下，比如隧道或者地铁中，没有GPS信号，就会出现无法识别的情况，并且现有技术中的可穿戴设备单次比对运动参数与状态阈值，判断运动状态不稳定，上报运动状态到监控终端时容易出现运动状态误判的现象。
发明内容
为解决上述技术问题，本发明提供一种运动状态识别方法、系统、可穿戴设备和存储介质，通过设计合加速度的特征值包括：均值、方差、峰度，将其输入决策树模型，根据决策结果判断目标运动状态，解决了现有技术中当GPS信号过弱时无法准确识别目标运动状态、以及在识别运动状态过程中误判的问题。提高了识别目标运动状态的准确度，减少了干扰信息对监控终端接收目标的运动状态的影响。
为实现本发明以上的发明目的，本发明是通过以下技术方案实现的：
本发明提供一种运动状态识别方法，包括：
采集目标的三维加速度数据；
根据所述目标的三维加速度数据，计算所述目标的合加速度；
提取所述目标的合加速度的特征值，所述合加速度的特征值包括：均值、方差、峰度；
将所述合加速度的特征值输入预设决策树模型，获得决策结果；
根据所述决策结果判断所述目标的当前运动状态。
该技术方案可以解决在GPS定位不准确或者GPS信号不好时如何通过测算目标的加速度来识别目标的运动状态的问题。通过提取目标合加速度的特征值包括：均值、方差、峰度，将特征值输入决策树模型并对决策结果的处理，可以准确的识别目标当前的运动状态，利用目标合加速度的均值、方差、峰度等特征值对区分不同运动状态的独特优势，可以有效地减小识别过程中的误差。
进一步地，所述的一种运动状态识别方法，还包括：
将单位时间内所述目标的若干个所述合加速度的特征值输入预设决策树模型，获得若干个决策结果；
对单位时间内所述若干个决策结果进行统计，得到单位时间内所述目标的当前运动状态；
进一步地，所述的一种运动状态识别方法中，所述对单位时间内所述若干个决策结果进行统计，得到单位时间内所述目标的当前运动状态，具体包括：
在单位时间内，对所述若干个决策结果进行统计，当所述目标的某一运动状态对应的若干个决策结果的数量占比达所述目标的若干个决策结果的总数量的预设阈值以上时，则判断所述目标当前处于所述运动状态；
在单位时间内，对所述若干个决策结果进行统计，当所述目标的某一运动状态对应的若干个决策结果的数量占比在所述目标的若干个决策结果的总数量的预设阈值以下时，则判断所述目标当前不处于所述运动状态。
进一步地，所述的一种运动状态识别方法，还包括：
在单位时间内，判断所述目标当前处于某一运动状态时，输出所述目标当前处于所述运动状态的信息；
在单位时间内，判断所述目标当前不处于某一运动状态时，不输出所述目标当前处于某一运动状态的信息。
进一步地，所述的一种运动状态识别方法，还包括：
在单位时间内，所述目标的任一运动状态对应的若干个决策结果的数量占比均在所述目标的若干个决策结果的总数量的预设阈值以下时，不输出所述目标当前运动状态的信息。
本发明还提供一种运动状态识别系统，包括：
采集模块，用于采集目标的三维加速度数据；
运算模块，用于根据所述目标的三维加速度数据，计算所述目标的合加速度；
提取模块，用于提取所述目标的合加速度的特征值，所述合加速度的特征值包括：均值、方差、峰度；
决策树模块，用于将所述合加速度的特征值输入预设决策树模型，获得决策结果；
统算模块，用于根据所述决策结果判断所述目标的当前运动状态；
通信模块，用于输出所述单位时间内的所述目标的当前运动状态的信息；
监控终端，用于接收所述单位时间内的所述目标的当前运动状态的信息。
进一步地，所述的一种运动状态识别系统中，所述统算模块包括：
将单位时间内所述目标的若干个所述合加速度的特征值输入预设决策树模型，获得若干个决策结果；
在单位时间内，对所述若干个决策结果进行统计，当所述目标的某一运动状态对应的若干个决策结果的数量占比达所述目标的若干个决策结果的总数量的预设阈值以上时，则判断所述目标当前处于所述运动状态；
在单位时间内，对所述若干个决策结果进行统计，当所述目标的某一运动状态对应的若干个决策结果的数量占比在所述目标的若干个决策结果的总数量的预设阈值以下时，则判断所述目标当前不处于所述运动状态。
进一步地，所述的一种运动状态识别系统中，所述通信模块包括：
在单位时间内，判断所述目标当前处于某一运动状态时，输出所述目标当前处于所述运动状态的信息；
在单位时间内，判断所述目标当前不处于某一运动状态时，不输出所述目标当前处于某一运动状态的信息；
在单位时间内，所述目标的任一运动状态对应的若干个决策结果的数量占比均在所述目标的若干个决策结果的总数量的预设阈值以下时，不输出所述目标当前运动状态的信息。
本发明还提供一种可穿戴设备，包括加速度传感器、处理器、储存器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，所述处理器，用于执行所述存储器上所存放的计算机程序，实现如权利要求1至权利要求5任一项所述的运动状态识别方法所执行的操作。
本发明还提供一种存储介质，其特征在于，所述存储介质中存储有至少一条指令，所述指令由处理器加载并执行以实现如权利要求1至权利要求5任一项所述的运动状态识别方法所执行的操作。
本发明提供一种运动状态识别方法、系统、可穿戴设备和存储介质，至少具有以下增益效果：
1)、减少可穿戴设备识别运动状态过程中对于GPS定位的依赖，通过加速度传感器采集的目标加速度，提取目标合加速度的特征值，即可完成对目标运动状态的识别；
2)、通过分析目标合加速度中的特征值包括均值、方差、峰度，将上述 目标合加速度的特征值输入决策树模型进行决策，根据决策结果完成对目标当前运动状态的识别，可以减少运动状态识别的误差；
3)、将目标运动状态对应的决策结果统计后与预设阈值进行比对，判断目标当前的运动状态，可以提高目标当前的运动状态识别的精确度；
4)、当判断目标当前的运动状态精确度不高时，不将目标当前的运动状态的信息输出给监控终端，提高了监控终端获取状态信息的准确性。
下面将以明确易懂的方式，结合附图说明优选实施方式，对一种充电电池组件防过放的保护方法、系统、装置的上述特性、技术特征、优点及其实现方式予以进一步说明。
图1是本发明一种运动状态识别方法的一个实施例的流程图；
图2是本发明一种运动状态识别方法中决策树的一个实施例的流程图；
图3是本发明一种运动状态识别方法的另一实施例的流程图；
图4是本发明一种运动状态识别方法的另一实施例的流程图；
图5是本发明一种运动状态识别系统的一个实施例的流程图；
图6时本发明一种可穿戴装置的一个实施例的结构示意图；
图中标号：10-采集模块、20-运算模块、30-提取模块、40-决策树模块、50-统算模块、60-通信模块、70-监控终端、100-可穿戴设备、110-加速度传感器、120-存储器、121-计算机程序、130-处理器。
以下描述中，为了说明而不是为了限定，提出了诸如特定系统结构、技术之类的具体细节，以便透彻理解本申请实施例。然而，本领域的技术人员应当清楚，在没有这些具体细节的其他实施例中也可以实现本申请。在其他情况中， 省略对众所周知的系统、装置、电路以及方法的详细说明，以免不必要的细节妨碍本申请的描述。
应当理解，当在本说明书和所附权利要求书中使用时，术语“包括”指示所述描述特征、整体、步骤、操作、元素和/或组件的存在，但并不排除一个或多个其他特征、整体、步骤、操作、元素、组件和/或集合的存在或添加。
为使图面简洁，各图中只示意性地表示出了与本发明相关的部分，它们并不代表其作为产品的实际结构。另外，以使图面简洁便于理解，在有些图中具有相同结构或功能的部件，仅示意性地绘示了其中的一个，或仅标出了其中的一个。在本文中，“一个”不仅表示“仅此一个”，也可以表示“多于一个”的情形。
还应当进一步理解，在本申请说明书和所附权利要求书中使用的术语“和/或”是指相关联列出的项中的一个或多个的任何组合以及所有可能组合，并且包括这些组合。
另外，在本申请的描述中，术语“第一”、“第二”等仅用于区分描述，而不能理解为指示或暗示相对重要性。
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对照附图说明本发明的具体实施方式。显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图，并获得其他的实施方式。
本发明的一个实施例，如图1所示，一种运动状态识别方法，包括：
S100采集目标的三维加速度数据；
具体地，采集目标的三维加速度数据是指采集目标的加速度传感器x、y、z三个方向轴上的加速度a
x、a
y、a
z。
S200根据目标的三维加速度数据，计算目标的合加速度；
具体地，采集目标的三维加速度数据a
x、a
y、a
z,考虑到可穿戴设备形式以及动作的多样性,三个方向轴分开进行加速度特征提取意义不大,因此选择先将目标的加速度进行合并，提取三个方向轴上的加速度a
x、a
y、a
z的矢量和为合加速度a，再对合加速度a进行特征提取.其中合加速度a计算为：
S300提取目标的合加速度特征值，合加速度的特征值包括：均值、方差、峰度；
具体地，提取特征值中均值是表征目标在单位时间内的加速度平均量，当目标处于加速度不同的运动状态下时，可以用均值来区分目标不同的运动状态，例如靠走路状态、跑步状态、乘轿车状态、乘地铁状态、乘公交车状态的加速度均值不同，区分目标处于不同运动状态；
提取特征值中的方差是表征目标在单位时间内的加速度离散程度，当目标处于加速度离散程度不同的运动状态下时，也可以用来区分目标不同的运动状态，例如匀速走路运动状态时单位时间内加速度离散程度低，当乘地铁运动状态时由于存在到站停车的情况，所以单位时间内加速度离散程度高，同时利用方差还可以减少对目标运动状态识别的误差，其中方差的计算为：
提取特征值中的峰度是表征目标在单位时间内的加速度表征概率分布密度函数曲线在平均值处的状态，可以通过对峰度的判断，减少对目标运动状态识别的误差，提高对运动状态识别的精确度，其中峰度的计算为：
可选地，在提取合加速度特征值时，还可以提取合加速度的偏度。提取特征值中的偏度是表征目标在单位时间内的概率分布密度函数曲线相对于平均值的不对称程度，可以通过对偏度的判断，减少对目标运动状态识别的误差， 提高对运动状态识别的精确度，其中偏度的计算为：
S400将合加速度的特征值输入预设决策树模型，获得决策结果；
具体地，预设决策树模型建立的步骤为：
建立基础模型；
采集训练数据；
将训练数据输入基础模型中，得到最终决策树模型。
将提取出来的加速度的特征值输入决策树模型进行判别，其中决策树模型中的参数由12个小时以上的测试数据训练而来的。
示例性地，如图2所示，通过举例说明将合加速度特征值输入某一决策树模型后，获得决策结果的详细判断过程包括：
将合加速度的特征值输入决策树模型，当第一次决策合加速度的特征值的方差小于第一方差值时，进入第二次决策，合加速度的均值小于第一均值时，输出判定结果为B；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值小于第一方差值时，进入第二次决策，合加速度的均值大于第一均值时，进入第三次决策，当合加速度的方差值小于第三方差值时，输出决策结果为C；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值小于第一方差值时，进入第二次决策，合加速度的均值大于第一均值时，进入第三次决策，当合加速度的方差值大于第三方差值时，输出决策结果为A；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值小于第二方差值时，进入第三次决策，当合加速度的峰度小于第一峰度时，输出决策结果为A；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大 于第一方差值时，进入第二次决策，当合加速度的方差值小于第二方差值时，进入第三次决策，当合加速度的峰度大于第一峰度时，进入第四次决策，当合加速度的均值大于第二均值时，输出决策结果为A；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值小于第二方差值时，进入第三次决策，当合加速度的峰度大于第一峰度时，进入第四次决策，当合加速度的均值小于于第二均值时，进入第五次决策，当方差小于第三方差值时，输出决策结果为B；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值小于第二方差值时，进入第三次决策，当合加速度的峰度大于第一峰度时，进入第四次决策，当合加速度的均值小于于第二均值时，进入第五次决策，当合加速度的方差大于第三方差值时，进入第六次决策，当合加速度的峰度小于第三峰度时，输出决策结果为A；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值小于第二方差值时，进入第三次决策，当合加速度的峰值大于第一峰度时，进入第四次决策，当合加速度的均值小于于第二均值时，进入第五次决策，当合加速度的方差大于第三方差值时，进入第六次决策，当合加速度的峰度大于第三峰值时，输出决策结果为B；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值大于第二方差值时，进入第三次决策，当合加速度的峰度大于第二峰度时，输出决策结果为B；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值大于第二方差值时， 进入第三次决策，当合加速度的峰度小于于第二峰度时，进入第三次决策，当合加速度的均值大于第三均值时，输出决策结果为C；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值大于第二方差值时，进入第三次决策，当合加速度的峰度小于于第二峰度时，进入四次决策，当合加速度的均值小于第三均值时，进入第五次决策，当合加速度的方差小于第四方差值时，输出决策结果为C；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值大于第二方差值时，进入第三次决策，当合加速度的峰度小于于第二峰度时，进入四次决策，当合加速度的均值小于第三均值时，进入第五次决策，当合加速度的方差大于第四方差值时，输出决策结果为B。
S500根据决策结果判断所述目标的当前运动状态。
本实施例可以解决在GPS定位不准确或者GPS信号不好时如何通过测算目标的加速度来识别目标的运动状态的问题。通过提取目标合加速度的特征值包括：均值、方差、峰度，将特征值输入决策树模型并对决策结果的处理，可以准确的识别目标当前的运动状态，利用目标合加速度的均值、方差、峰度等特征值对区分不同运动状态的独特优势，可以有效地减小识别过程中的误差。
本发明的另一个实施例，如图3所示，一种运动状态识别方法，包括：
S100采集目标的三维加速度数据；
具体地，采集目标的三维加速度数据是指采集目标的加速度传感器x、y、z三个方向轴上的加速度a
x、a
y、a
z。
S200根据目标的三维加速度数据，计算目标的合加速度；
具体地，采集目标的三维加速度数据a
x、a
y、a
z,考虑到可穿戴设备形式以 及动作的多样性,三个方向轴分开进行加速度特征提取意义不大,因此选择先将目标的加速度进行合并，提取三个方向轴上的加速度a
x、a
y、a
z的矢量和为合加速度a，再对合加速度a进行特征提取.其中合加速度a计算为：
S300提取目标的合加速度特征值，合加速度的特征值包括：均值、方差、峰度；
具体地，提取特征值中均值是表征目标在单位时间内的加速度平均量，当目标处于加速度不同的运动状态下时，可以用均值来区分目标不同的运动状态，例如靠走路状态、跑步状态、乘轿车状态、乘地铁状态、乘公交车状态的加速度均值不同，区分目标处于不同运动状态；
提取特征值中的方差是表征目标在单位时间内的加速度离散程度，当目标处于加速度离散程度不同的运动状态下时，也可以用来区分目标不同的运动状态，例如匀速走路运动状态时单位时间内加速度离散程度低，当乘地铁运动状态时由于存在到站停车的情况，所以单位时间内加速度离散程度高，同时利用方差还可以减少对目标运动状态识别的误差，其中方差的计算为：
提取特征值中的峰度是表征目标在单位时间内的加速度表征概率分布密度函数曲线在平均值处的状态，可以通过对峰度的判断，减少对目标运动状态识别的误差，提高对运动状态识别的精确度，其中峰度的计算为：
可选地，在提取合加速度特征值时，还可以提取合加速度的偏度。提取特征值中的偏度是表征目标在单位时间内的概率分布密度函数曲线相对于平均值的不对称程度，可以通过对偏度的判断，减少对目标运动状态识别的误差，提高对运动状态识别的精确度，其中偏度的计算为：
S400将合加速度的特征值输入预设决策树模型，获得决策结果；
具体地，预设决策树模型建立的步骤为：
建立基础模型；
采集训练数据；
将训练数据输入基础模型中，得到最终决策树模型。
将提取出来的加速度的特征值输入决策树模型进行判别，其中决策树模型中的参数由12个小时以上的测试数据训练而来的。
示例性地，如图2所示，通过举例说明将合加速度特征值输入某一决策树模型后，获得决策结果的详细判断过程包括：
将合加速度的特征值输入决策树模型，当第一次决策合加速度的特征值的方差小于第一方差值时，进入第二次决策，合加速度的均值小于第一均值时，输出判定结果为B；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值小于第一方差值时，进入第二次决策，合加速度的均值大于第一均值时，进入第三次决策，当合加速度的方差值小于第三方差值时，输出决策结果为C；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值小于第一方差值时，进入第二次决策，合加速度的均值大于第一均值时，进入第三次决策，当合加速度的方差值大于第三方差值时，输出决策结果为A；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值小于第二方差值时，进入第三次决策，当合加速度的峰度小于第一峰度时，输出决策结果为A；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值小于第二方差值时， 进入第三次决策，当合加速度的峰度大于第一峰度时，进入第四次决策，当合加速度的均值大于第二均值时，输出决策结果为A；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值小于第二方差值时，进入第三次决策，当合加速度的峰度大于第一峰度时，进入第四次决策，当合加速度的均值小于于第二均值时，进入第五次决策，当方差小于第三方差值时，输出决策结果为B；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值小于第二方差值时，进入第三次决策，当合加速度的峰度大于第一峰度时，进入第四次决策，当合加速度的均值小于于第二均值时，进入第五次决策，当合加速度的方差大于第三方差值时，进入第六次决策，当合加速度的峰度小于第三峰度时，输出决策结果为A；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值小于第二方差值时，进入第三次决策，当合加速度的峰值大于第一峰度时，进入第四次决策，当合加速度的均值小于于第二均值时，进入第五次决策，当合加速度的方差大于第三方差值时，进入第六次决策，当合加速度的峰度大于第三峰值时，输出决策结果为B；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值大于第二方差值时，进入第三次决策，当合加速度的峰度大于第二峰度时，输出决策结果为B；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值大于第二方差值时，进入第三次决策，当合加速度的峰度小于于第二峰度时，进入第三次决策，当 合加速度的均值大于第三均值时，输出决策结果为C；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值大于第二方差值时，进入第三次决策，当合加速度的峰度小于于第二峰度时，进入四次决策，当合加速度的均值小于第三均值时，进入第五次决策，当合加速度的方差小于第四方差值时，输出决策结果为C；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值大于第二方差值时，进入第三次决策，当合加速度的峰度小于于第二峰度时，进入四次决策，当合加速度的均值小于第三均值时，进入第五次决策，当合加速度的方差大于第四方差值时，输出决策结果为B。
S510在单位时间内，对若干个决策结果进行统计，当目标的某一运动状态对应的决策结果的数量占比达目标的若干个决策结果的总数量的预设阈值以上时，则判断该运动状态为目标的当前运动状态。
具体地，预设阈值可以设置在50％-90％之间，当预设阈值设置在80％时，即当目标的某一运动状态对应的决策结果的数量占比在目标的若干个决策结果的总数量的80％以上时，则判断该运动状态为目标的当前运动状态。
在本实施例中，具体地说明如何根据单位时间内的若干个决策结果判断所述目标的当前运动状态，通过预设阈值的设计，减少在识别目标运动状态过程中产生的误差，提高运动状态识别的准确性。
本发明的另一个实施例，如图4所示，在前一实施例的基础上，一种运动状态识别方法，还包括：
S610判断某一运动状态为目标的当前运动状态之后，输出目标的当前运动状态信息；
S620在单位时间内，目标的任一运动状态对应的决策结果的数量占比均在目标的若干个决策结果的总数量的预设阈值以下时，不输出目标的运动状态的信息；
在本实施例中，当判断目标当前运动状态精确度不高时，通信模块不将目标当前的运动状态的信息输出给监控终端，提高了监控终端获取状态信息的准确性。
本发明的一个实施例，如图5所示，本发明还提供一种运动状态识别系统，包括：
采集模块10，用于采集目标的三维加速度数据；
具体地，采集目标的三维加速度数据是指采集目标的加速度传感器x、y、z三个方向轴上的加速度a
x、a
y、a
z。
运算模块20，用于根据所述目标的三维加速度数据，计算所述目标的合加速度；
具体地，采集目标的三维加速度数据a
x、a
y、a
z,考虑到可穿戴设备形式以及动作的多样性,三个方向轴分开进行加速度特征提取意义不大,因此选择先将目标的加速度进行合并，提取三个方向轴上的加速度a
x、a
y、a
z的矢量和为合加速度a，再对合加速度a进行特征提取.其中合加速度a计算为：
提取模块30，用于提取所述目标的合加速度的特征值，所述合加速度的特征值包括：均值、方差、峰度；
具体地，提取特征值中均值是表征目标在单位时间内的加速度平均量，当目标处于加速度不同的运动状态下时，可以用均值来区分目标不同的运动状态，例如靠走路状态、跑步状态、乘轿车状态、乘地铁状态、乘公交车状态的加速度均值不同，区分目标处于不同运动状态；
提取特征值中的方差是表征目标在单位时间内的加速度离散程度，当目标处于加速度离散程度不同的运动状态下时，也可以用来区分目标不同的运动状态，例如匀速走路运动状态时单位时间内加速度离散程度低，当乘地铁运动状态时由于存在到站停车的情况，所以单位时间内加速度离散程度高，同时利用方差还可以减少对目标运动状态识别的误差，其中方差的计算为：
提取特征值中的峰度是表征目标在单位时间内的加速度表征概率分布密度函数曲线在平均值处的状态，可以通过对峰度的判断，减少对目标运动状态识别的误差，提高对运动状态识别的精确度，其中峰度的计算为：
可选地，在提取合加速度特征值时，还可以提取合加速度的偏度。提取特征值中的偏度是表征目标在单位时间内的概率分布密度函数曲线相对于平均值的不对称程度，可以通过对偏度的判断，减少对目标运动状态识别的误差，提高对运动状态识别的精确度，其中偏度的计算为：
决策树模块40，用于将所述合加速度的特征值输入预设决策树模型，获得决策结果；
具体地，预设决策树模型建立的步骤为：
建立基础模型；
采集训练数据；
将训练数据输入基础模型中，得到最终决策树模型。
将提取出来的加速度的特征值输入决策树模型进行判别，其中决策树模型中的参数由12个小时以上的测试数据训练而来的。
示例性地，如图2所示，通过举例说明将合加速度特征值输入某一决策树模型后，获得决策结果的详细判断过程包括：
将合加速度的特征值输入决策树模型，当第一次决策合加速度的特征值的方差小于第一方差值时，进入第二次决策，合加速度的均值小于第一均值时，输出判定结果为B；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值小于第一方差值时，进入第二次决策，合加速度的均值大于第一均值时，进入第三次决策，当合加速度的方差值小于第三方差值时，输出决策结果为C；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值小于第一方差值时，进入第二次决策，合加速度的均值大于第一均值时，进入第三次决策，当合加速度的方差值大于第三方差值时，输出决策结果为A；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值小于第二方差值时，进入第三次决策，当合加速度的峰度小于第一峰度时，输出决策结果为A；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值小于第二方差值时，进入第三次决策，当合加速度的峰度大于第一峰度时，进入第四次决策，当合加速度的均值大于第二均值时，输出决策结果为A；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值小于第二方差值时，进入第三次决策，当合加速度的峰度大于第一峰度时，进入第四次决策，当合加速度的均值小于于第二均值时，进入第五次决策，当方差小于第三方差值时，输出决策结果为B；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值小于第二方差值时， 进入第三次决策，当合加速度的峰度大于第一峰度时，进入第四次决策，当合加速度的均值小于于第二均值时，进入第五次决策，当合加速度的方差大于第三方差值时，进入第六次决策，当合加速度的峰度小于第三峰度时，输出决策结果为A；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值小于第二方差值时，进入第三次决策，当合加速度的峰值大于第一峰度时，进入第四次决策，当合加速度的均值小于于第二均值时，进入第五次决策，当合加速度的方差大于第三方差值时，进入第六次决策，当合加速度的峰度大于第三峰值时，输出决策结果为B；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值大于第二方差值时，进入第三次决策，当合加速度的峰度大于第二峰度时，输出决策结果为B；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值大于第二方差值时，进入第三次决策，当合加速度的峰度小于于第二峰度时，进入第三次决策，当合加速度的均值大于第三均值时，输出决策结果为C；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值大于第二方差值时，进入第三次决策，当合加速度的峰度小于于第二峰度时，进入四次决策，当合加速度的均值小于第三均值时，进入第五次决策，当合加速度的方差小于第四方差值时，输出决策结果为C；
将合加速度的特征值输入决策树模型，当第一次决策合加速度的方差值大于第一方差值时，进入第二次决策，当合加速度的方差值大于第二方差值时，进入第三次决策，当合加速度的峰度小于于第二峰度时，进入四次决策，当合 加速度的均值小于第三均值时，进入第五次决策，当合加速度的方差大于第四方差值时，输出决策结果为B。
统算模块50，用于根据决策结果判断目标的当前运动状态；
通信模块60，用于输出单位时间内的目标的当前运动状态的信息；
监控终端70，用于接收单位时间内的目标的当前运动状态的信息。
本实施例可以解决在GPS定位不准确或者GPS信号不好时如何通过测算目标的加速度来识别目标的运动状态的问题。通过提取目标合加速度的特征值包括：均值、方差、峰度，将特征值输入决策树模型并对决策结果的处理，可以准确的识别目标当前的运动状态，利用目标合加速度的均值、方差、峰度等特征值对区分不同运动状态的独特优势，可以有效地减小识别过程中的误差。
基于前述实施例，统算模块具体还包括以下功能：
在单位时间内，对若干个决策结果进行统计，当目标的某一运动状态对应的决策结果的数量占比达目标的若干个决策结果的总数量的预设阈值以上时，则判断该运动状态为目标的当前运动状态。
具体地，预设阈值可以设置在50％-90％之间，当预设阈值设置在80％时，即当目标的某一运动状态对应的若干个决策结果的数量占比在目标的若干个决策结果的总数量的80％以上时，则判断目标当前处于该运动状态；当目标的某一运动状态对应的若干个决策结果的数量占比在目标的若干个决策结果的总数量的80％以下时，则判断目标当前不处于该运动状态。
在本模块中，具体地说明如何根据单位时间内的若干个决策结果判断所述目标的当前运动状态，通过预设阈值的设计，减少在识别目标运动状态过程中产生的误差，提高运动状态识别的准确性。
基于前述实施例，通信模块还包括以下功能：
判断该运动状态为目标的当前运动状态之后，输出目标的当前运动状态信息；
在单位时间内，目标的任一运动状态对应的决策结果的数量占比均在目标的若干个决策结果的总数量的预设阈值以下时，不输出目标的运动状态的信息。
在本模块中，当判断目标当前运动状态精确度不高时，通信模块不将目标当前的运动状态的信息输出给监控终端，提高了监控终端获取状态信息的准确性。
本发明的一个实施例，如图6所示，本发明还提供一种可穿戴设备100，包括加速度传感器110、存储器120，用于存放计算机程序121、处理器130，用于执行存储器120上所存放的计算机程序121，实现上述所对应方法实施例中的识别运动状态方法。
可穿戴设备100可以为桌上型计算机、笔记本、掌上电脑、平板型计算机、手机、人机交互屏等设备。可穿戴设备100可包括，但不仅限于处理器130、存储器120。本领域技术人员可以理解，图6仅仅是可穿戴100的示例，并不构成对可穿戴设备100的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件，例如：可穿戴设备100还可以包括输入/输出接口、显示设备、网络接入设备、通信总线、通信接口等。通信接口和通信总线，还可以包括输入/输出接口，其中，处理器130、存储器120、输入/输出接口和通信接口通过通信总线完成相互间的通信。该存储器120存储有计算机程序121，该处理器130用于执行存储器120上所存放的计算机程序121，实现上述所对应方法实施例中的拍照控制方法。
处理器130可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、 专用集成电路(Application Specific Integrated Circuit，ASIC)、现场可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
存储器120可以是可穿戴设备100的内部存储单元，例如：可穿戴设备的硬盘或内存。存储器也可以是所可穿戴设备的外部存储设备，例如：可穿戴设备上配备的插接式硬盘，智能存储卡(Smart Media Card，SMC)，安全数字(Secure Digital，SD)卡，闪存卡(Flash Card)等。进一步地，存储器120还可以既包括可穿戴设备100的内部存储单元也包括外部存储设备。存储器120用于存储计算机程序121以及可穿戴设备100所需要的其他程序和数据。可穿戴设备还可以用于暂时地存储已经输出或者将要输出的数据。
通信总线是连接所描述的元素的电路并且在这些元素之间实现传输。例如，处理器130通过通信总线从其它元素接收到命令，解密接收到的命令，根据解密的命令执行计算或数据处理。存储器120可以包括程序模块，例如内核(kernel)，中间件(middleware)，应用程序编程接口(Application Programming Interface，API)和应用。该程序模块可以是有软件、固件或硬件、或其中的至少两种组成。输入/输出接口转发用户通过输入/输出接口(例如感应器、键盘、触摸屏)输入的命令或数据。通信接口将该可穿戴设备100与其它网络设备、用户设备、网络进行连接。例如，通信接口可以通过有线或无线连接到网络以连接到外部其它的网络设备或用户设备。无线通信可以包括以下至少一种：无线保真(WiFi)，蓝牙(BT)，近距离无线通信技术(NFC)，全球卫星定位系统(GPS)和蜂窝通信等等。有线通信可以包括以下至少一种：通用串行总线(USB)，高清晰度多媒体接口(HDMI)，异步传输标准接口(RS-232)等等。网络可以是电信网络和通信网络。通信网络可以为计算机网络、因特网、物联网、电话网络。可穿戴设备100可以通过通信接口连接网络，可穿戴设备100和其它网络设备通 信所用的协议可以被应用、应用程序编程接口(API)、中间件、内核和通信接口至少一个支持。
本发明的一个实施例，本发明还提供一种存储介质，存储介质中存储有至少一条指令，指令由处理器加载并执行以实现上述拍照控制方法对应实施例所执行的操作。例如，存储介质可以是只读内存(ROM)、随机存取存储器(RAM)、只读光盘(CD-ROM)、磁带、软盘和光数据存储设备等。
它们可以用计算装置可执行的程序代码来实现，从而，可以将它们存储在存储装置中由计算装置来执行，或者将它们分别制作成各个集成电路模块，或者将它们中的多个模块或步骤制作成单个集成电路模块来实现。这样，本发明不限制于任何特定的硬件和软件结合。
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详细描述或记载的部分，可以参见其他实施例的相关描述。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
在本申请所提供的实施例中，应该理解到，所揭露的装置/智能手表和方法，可以通过其他的方式实现。例如，以上所描述的装置/智能手表实施例仅仅是示意性的，例如，所述模块或单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如，多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通讯连接可以是通过一些接口，装置或单元的间接耦合或通讯连接，可以是电性、机械或其他的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可能集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
应当说明的是，上述实施例均可根据需要自由组合。以上所述仅是本发明的优选实施方式，应当指出，对于本技术领域的普通技术人员来说，在不脱离本发明原理的前提下，还可以做出若干改进和润饰，这些改进和润饰也应视为本发明的保护范围。
Claims (10)
- 一种运动状态识别方法，其特征在于，包括：采集目标的三维加速度数据；根据所述目标的三维加速度数据，计算所述目标的合加速度；提取所述目标的合加速度的特征值，所述合加速度的特征值包括：均值、方差、峰度；将所述合加速度的特征值输入预设决策树模型，获得决策结果；根据所述决策结果判断所述目标的当前运动状态。
- 根据权利要求1所述的一种运动状态识别方法，其特征在于，还包括：将单位时间内所述目标的若干个所述合加速度的特征值输入预设决策树模型，获得若干个所述决策结果；对单位时间内若干个所述决策结果进行统计，得到单位时间内所述目标的当前运动状态；
- 根据权利要求2所述的一种运动状态识别方法，其特征在于，所述对单位时间内的若干个所述决策结果进行统计，得到单位时间内所述目标的当前运动状态，具体包括：在单位时间内，对若干个所述决策结果进行统计，当所述目标的某一运动状态对应的若干个决策结果的数量占比达所述目标的若干个决策结果的总数量的预设阈值以上时，则判断该运动状态为所述目标的当前运动状态。
- 根据权利要求3所述的一种运动状态识别方法，其特征在于，所述判断该运动状态为所述目标的当前运动状态之后，还包括步骤：输出所述目标的当前运动状态信息。
- 根据权利要求3所述的一种运动状态识别方法，其特征在于，还包括：在单位时间内，所述目标的任一运动状态对应的决策结果的数量占比均在所述目标的若干个决策结果的总数量的预设阈值以下时，不输出所述目标的运动状态的信息。
- 一种运动状态识别系统，其特征在于，包括：采集模块，用于采集目标的三维加速度数据；运算模块，用于根据所述目标的三维加速度数据，计算所述目标的合加速度；提取模块，用于提取所述目标的合加速度的特征值，所述合加速度的特征值包括：均值、方差、峰度；决策树模块，用于将所述合加速度的特征值输入预设决策树模型，获得决策结果；统算模块，用于根据所述决策结果判断所述目标的当前运动状态；通信模块，用于输出所述单位时间内的所述目标的当前运动状态的信息；监控终端，用于接收所述单位时间内的所述目标的当前运动状态的信息。
- 根据权利要求6所述的一种运动状态识别系统，其特征在于，所述统算模块包括：将单位时间内所述目标的若干个所述合加速度的特征值输入预设决策树模型，获得若干个决策结果；在单位时间内，对若干个所述决策结果进行统计，当所述目标的某一运动状态对应的决策结果的数量占比达所述目标的若干个决策结果的总数量的预设阈值以上时，则判断该运动状态为所述目标的当前运动状态；
- 根据权利要求7所述的一种运动状态识别系统，其特征在于，所述通信模块包括：所述统算模块判断所述运动状态为所述目标的当前运动状态后，输出所述目标的当前运动状态信息；所述统算模块判断所述目标的任一运动状态对应的决策结果的数量占比均在所述目标的若干个决策结果的总数量的预设阈值以下时，不输出所述目标的运动状态的信息。
- 一种可穿戴设备，其特征在于，包括加速度传感器、处理器、储存器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，所述处理器，用于执行所述存储器上所存放的计算机程序，实现如权利要求1至权利要求5任一项所述的运动状态识别方法所执行的操作。
- 一种存储介质，其特征在于，所述存储介质中存储有至少一条指令，所述指令由处理器加载并执行以实现如权利要求1至权利要求5任一项所述的运动状态识别方法所执行的操作。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110481572.6 | 2021-04-30 | ||
| CN202110481572.6A CN113008231A (zh) | 2021-04-30 | 2021-04-30 | 一种运动状态识别方法、系统、可穿戴设备和存储介质 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2022227202A1 true WO2022227202A1 (zh) | 2022-11-03 |
Family
ID=76380534
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2021/097181 WO2022227202A1 (zh) | 2021-04-30 | 2021-05-31 | 一种运动状态识别方法、系统、可穿戴设备和存储介质 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113008231A (zh) |
| WO (1) | WO2022227202A1 (zh) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114385012B (zh) * | 2022-01-17 | 2023-06-30 | 维沃移动通信有限公司 | 运动的识别方法、装置、电子设备和可读存储介质 |
| CN114444788A (zh) * | 2022-01-17 | 2022-05-06 | 北京卡路里信息技术有限公司 | Ftp的预测方法和装置、存储介质及处理器 |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160262685A1 (en) * | 2013-11-12 | 2016-09-15 | Highland Instruments, Inc. | Motion analysis systemsand methods of use thereof |
| CN107220617A (zh) * | 2017-05-25 | 2017-09-29 | 哈尔滨工业大学 | 人体姿态识别系统及方法 |
| CN107462258A (zh) * | 2017-07-13 | 2017-12-12 | 河海大学 | 一种基于手机三轴加速度传感器的计步方法 |
| CN108009572A (zh) * | 2017-11-22 | 2018-05-08 | 中国地质大学（武汉） | 移动设备坠落检测方法及其模型形成方法与移动设备 |
| CN108898062A (zh) * | 2018-05-31 | 2018-11-27 | 电子科技大学 | 一种基于改进的信号片段提取算法的手部动作识别方法 |
| CN109166624A (zh) * | 2018-09-21 | 2019-01-08 | 广州杰赛科技股份有限公司 | 一种行为分析方法、装置、服务器、系统及存储介质 |
| CN111401435A (zh) * | 2020-03-13 | 2020-07-10 | 安徽工业大学 | 一种基于运动手环的人体运动模式识别方法 |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012047756A1 (en) * | 2010-10-04 | 2012-04-12 | Avocent | System and method for monitoring and managing data center resources incorporating a common data model repository |
| CN106897659B (zh) * | 2015-12-18 | 2019-05-24 | 腾讯科技（深圳）有限公司 | 眨眼运动的识别方法和装置 |
| CN107582061B (zh) * | 2017-07-21 | 2020-03-27 | 青岛海信移动通信技术股份有限公司 | 一种识别人体运动状态的方法及智能移动设备 |
| CN108615050B (zh) * | 2018-04-11 | 2020-10-09 | 南京邮电大学 | 一种基于支持向量机的人体姿态判别方法 |
| CN108814618B (zh) * | 2018-04-27 | 2021-08-31 | 歌尔科技有限公司 | 一种运动状态的识别方法、装置及终端设备 |
| CN109558841B (zh) * | 2018-11-30 | 2023-06-02 | 歌尔科技有限公司 | 一种运动状态的识别方法、装置及终端 |
-
2021
- 2021-04-30 CN CN202110481572.6A patent/CN113008231A/zh active Pending
- 2021-05-31 WO PCT/CN2021/097181 patent/WO2022227202A1/zh active Application Filing
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160262685A1 (en) * | 2013-11-12 | 2016-09-15 | Highland Instruments, Inc. | Motion analysis systemsand methods of use thereof |
| CN107220617A (zh) * | 2017-05-25 | 2017-09-29 | 哈尔滨工业大学 | 人体姿态识别系统及方法 |
| CN107462258A (zh) * | 2017-07-13 | 2017-12-12 | 河海大学 | 一种基于手机三轴加速度传感器的计步方法 |
| CN108009572A (zh) * | 2017-11-22 | 2018-05-08 | 中国地质大学（武汉） | 移动设备坠落检测方法及其模型形成方法与移动设备 |
| CN108898062A (zh) * | 2018-05-31 | 2018-11-27 | 电子科技大学 | 一种基于改进的信号片段提取算法的手部动作识别方法 |
| CN109166624A (zh) * | 2018-09-21 | 2019-01-08 | 广州杰赛科技股份有限公司 | 一种行为分析方法、装置、服务器、系统及存储介质 |
| CN111401435A (zh) * | 2020-03-13 | 2020-07-10 | 安徽工业大学 | 一种基于运动手环的人体运动模式识别方法 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113008231A (zh) | 2021-06-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2022227202A1 (zh) | 一种运动状态识别方法、系统、可穿戴设备和存储介质 | |
| CN109189950B (zh) | 多媒体资源分类方法、装置、计算机设备及存储介质 | |
| CN108280477A (zh) | 用于聚类图像的方法和装置 | |
| US10909232B2 (en) | Terminal verification method, terminal device, and computer readable storage medium | |
| CN113280265B (zh) | 工况识别方法、装置、计算机设备以及存储介质 | |
| CN106503631A (zh) | 一种人群分析方法及计算机设备 | |
| CN111797148A (zh) | 数据处理方法、装置、存储介质及电子设备 | |
| CN113379469A (zh) | 一种异常流量检测方法、装置、设备及存储介质 | |
| CN110909804B (zh) | 基站异常数据的检测方法、装置、服务器和存储介质 | |
| CN108573218A (zh) | 人脸数据采集方法及终端设备 | |
| CN114333997A (zh) | 数据处理、数据处理模型的训练方法、装置、设备及介质 | |
| CN114356846A (zh) | 一种图像数据发送方法、装置、电子设备及存储介质 | |
| CN114298123A (zh) | 聚类处理方法、装置、电子设备及可读存储介质 | |
| CN110188602A (zh) | 视频中的人脸识别方法及装置 | |
| CN111506624B (zh) | 一种电力缺失数据辨识方法和相关装置 | |
| WO2018014432A1 (zh) | 语音应用触发控制方法、装置及终端 | |
| CN115665852B (zh) | 室内定位方法、装置、设备及存储介质 | |
| CN107832690B (zh) | 人脸识别的方法及相关产品 | |
| EP3940667A2 (en) | Vehicle association method and device, roadside equipment and cloud control platform | |
| CN113343709B (zh) | 意图识别模型的训练方法、意图识别方法、装置及设备 | |
| WO2019001253A1 (zh) | 虹膜活体检测方法及相关产品 | |
| CN115082202A (zh) | 用于房屋抵押的信息处理方法、装置、设备及存储介质 | |
| CN109213322B (zh) | 一种虚拟现实中手势识别的方法及系统 | |
| CN107995181B (zh) | 一种基于步态的身份验证方法、装置、设备及存储介质 | |
| CN112199226B (zh) | 问题确定方法及相关产品 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21938672 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 21938672 Country of ref document: EP Kind code of ref document: A1 |